Volunteer Information Exchange Sharing what we know with those we know Volume 1 Number 16 October 12, 2011

Questions Contribute To The VIE These questions need your answers When someone like Steve Jobs dies, Q: The Hollerith sorter has 26 slots. 24 of those are under there is so much to say, and so many control of the tabulator. Two have manual handles, and are people saying it, that it seems not be my not controlled by the tabulator. Tim Robinson asks, “Does place to try to add, except to say that he anyone know what those two manually operated slots will be missed. would have been for?” Many thanks to the several folks who Q: I know that when Xerox PARC gave extensive demos of responded to the ILLIAC articles. This the Alto , windows user interface, etc. to Xerox issue features the IBM SSEC. If you have executives in Rochester, NY, the execs were not stories about it, please them with impressed, but (some of) their wives were. My question is: us. I heard that one of those wives later started a high tech Do you have a favorite artifact, one that company. Who, what company, was it successful, and did you know a great deal about? One that they use anything from PARC? you know a great story about? Help us Kim Harris ensure that all those stories are passed along. Share your knowledge by contributing to the VIE. Jim Strickland [email protected] IN MEMORIAM

CONTENTS VIE 1 Open Questions 1 In Memoriam: Steve Jobs 1 vs. Aiken Round 2 SSEC 2 SSEC Pictures 3 Ted Codd and 3 Docent Quiz 4 “Don't Dress Like This!” 4 ILLIAC Revisited 5 Steve Jobs Twentieth Anniversary of Linux 5 1955 2011 Babbage Machine Repaired 6 the people who are crazy enough to they can Coming Events 6 change the world, are the ones who do. (Excerpt from an Apple “Think Different” advertisement)

Page 1 Watson vs. Aiken Round 2 – SSEC: The First Computer that Could Modify a Stored Program After the blowup with Howard Aiken in August of 1944, Thomas Watson Sr. reacted. He picked Frank Hamilton, who had been number two on the ASCC (Mark I) project with Aiken. "Put every IBM resource to work, Mr. Hamilton. I want the new machine to be faster than that one in Pennsylvania [ENIAC], more capable than the one we gave to Harvard. I want it to be installed here at World Headquarters, in one year. The SSEC Dedication Plaque And I want it to be available to the scientific world, not hidden away at a selfish university or at a military installation." He recruited Wallace Eckert (no relation to ENIAC's J. Presper Eckert) from Columbia University and put him in charge of the super calculator design. IBM spent $950,000 to develop the machine, named the Selective Sequence Electronic Calculator (SSEC). The SSEC could compute more than 250 times faster than Aiken's Mark I. It was the first computing machine that could he programmed with software. For more than four years, the SSEC fulfilled the wish that Watson had expressed at its dedication: Testing was completed in late 1947, when Watson that it would serve humanity by solving important made a decision that forever altered the public problems of science. perception of and linked IBM's name to the new generation of information machines. He told It enabled Wallace Eckert to publish a lunar the engineers to disassemble the SSEC and set it up ephemeris “... of greater accuracy than previously in the ground floor lobby of IBM's 590 Madison available... the source of data used in man's first Avenue (New York City) headquarters. The lobby landing on the moon". was open to the public, and its large external windows allowed a view of the SSEC for the information on cards and watched the results print multitudes cramming the sidewalks on Madison and out. 57th Street. There was a raised floor another world The first application of the SSEC was calculating the first. Watson would not have his guests, his positions of the moon and planets, known as the customers, his stockholders tripping over snarls of Ephemeris. Each position of the moon required cable. about 11,000 additions, 9,000 multiplications, and As part of putting the SSEC in front of the public, 2,000 table lookups, which took the SSEC about IBM announced that scientists could run problems seven minutes. This application used the machine through the SSEC for free, while commercial for about six months; by then other users were lined enterprises would pay $300 per hour, which was the up to keep the machine busy. cost of operating the machine. The point was to Pedestrians stopped to gawk and gave the SSEC keep the SSEC running, so anyone who looked in the nickname "Poppy." The New Yorker magazine would see that it worked. published a cover story on the SSEC and its public Watson told IBM designers to make sure the SSEC display. The machine influenced Hollywood, most looked sleek and impressive, just as Watson had famously as the model for the computer in the 1957 wanted the Mark I stylized. The spectacle of the movie Desk Set, featuring Katherine Hephurn and SSEC defined the public's image of a computer for Spencer Tracy. Before the SSEC, most people decades. Kept dustfree behind glass panels, reels thought of computers as fascinating but of tape ticked like clocks, punches stamped out incomprehensible laboratory experiments. Watson cards and whizzed them into hoppers, and took the computer out of the lab and sold it to the thousands of tiny lights flashed on and off in no public. As a public relations strategy, it worked discernible pattern. Operators standing at desksize brilliantly and more than consoles in the center of the room fed the SSEC made up for the Aiken fiasco. Continued on next page Page 2 Still, the SSEC was not a real electronic computer. Watson could not, or would not see the enormous The engineers had concocted a hybrid of electronics advantages of magnetic tape for data, such as the and old punch card machines. Information was storage space it would save, and the way it could stored on punch cards, not on a new and more shoot the data into a machine at great speed. As efficient invention: magnetic tape. Watson often pointed out, magnetic tape could Programs resided on paper tape which was actually accidentally get erased. Punch cards held uncut IBM card stock, more than seven inches wide, information on a tangible medium. weighing 400 pounds per roll. It remained for T. J. Watson Jr. to bring IBM into the electronic computer age.

The SSEC as installed at 590 Madison Avenue in New York City.

Watson Senior, upon first viewing SSEC prior to the public unveiling: "There is just one thing," he said somewhat offhandedly. "The sweep of this room is hindered by those large black columns down the center. Have them removed before the ceremony." But since they supported the building, the columns stayed. Instead, the photo in the brochure which was handed out at the ceremony, was carefully retouched to remove all traces of the offending columns. Note that the man at the rear is Ted Codd, then a programmer of the SSEC. Dr. Codd later became the father of Relational Data Base.

The retouched photo

Visible along the wall in the background are three punches and thirty readers that form the papertape storage, with a large roll of tape above each punch. The paper tape was actually uncut IBM card stock, more than seven inches wide, weighing 400 pounds per roll. Along the left wall are banks of circuits for card reading and sequence control and 36 paper tape readers comprising the tablelookup section, many of them loaded with custom tape loops for commonly referenced data. Most of the panels along the right wall are occupied by the electronic arithmetic unit and storage. In the center of the room: card readers, card punches, printers and the operator's console. Page 3 Ted Codd and John Backus: Early (SSEC). While on the tour, Backus mentioned to the Programmers of the SSEC guide that he was looking for a job. She encouraged Edgar, “Ted” Codd was a Britisher who flew for the him to talk to the director of the project and he was RAF during WW II. After the war, Codd moved to the hired to work on the SSEC. USA and in 1949 joined IBM as a mathematical The SSEC was not a computer in the modern sense. programmer for the SSEC, the huge vacuum tube It had no memory for software storage, and programs computer used to solve many of the largest scientific had to be entered on punched paper tape. It had problems of its day. thousands of electromechanical parts, making it In 1967, he moved to California to work at IBM's San unreliable and slow as well. Part of Backus's job was Jose Research Laboratory where he undertook the to attend the machine, and fix it when it would stop research that was to lead to the relational database running. Programming the SSEC was also a model. In 1981 Codd received the A M Turing challenge, as there was no set way of doing it. award, the highest honor in the field of computer Backus spent three years working on the SSEC, science. during which time he invented a program called As a matter of fact, Ted Codd is shown in the rear of Speedcoding. The program was the first to include a the “black column” photograph of the SSEC. scaling factor, which allowed both large and small numbers to be easily stored and manipulated. During the spring of 1949, John Backus visited the Backus went on to “invent” , the first and IBM Computer Center on Madison Avenue, where he arguably the most popular high level programming toured the Selective Sequence Electronic Calculator language.

Docent Quiz Shortly after I joined IBM in 1964, I went to my first IBM class. Docents: Take this test to be sure that you are qualified to work with our visitors. In addition to learning about equipment, (computer studies would come later) Spelling Test we learned about IBM's history and culture. a) How do you spell IBM? ______On an easel at the front of the class was a poster, b) How do you spell CDC? ______similar to the one below, showing the Beatles with c) How do you spell NCR? ______their soup bowl haircuts and collarless suits. Computer History Early IBM Tabulators used as input: a) punched cards b) punched cards c) punched cards d) all the above Word Scramble Rearrange the letters to make a word or words related to computer history. a) Large computer company: PH __ __ b) Desktop computer: CP __ __ c) World Wide Web: WWW: ______Museum Knowledge The 's address is: a) 1401 Shoreline b) PDP1 Shoreline c) 305 Ramac Under the poster was the admonishment Answers on page 6. “IBM'ers don't dress like this.” Jim Strickland

Page 4 The ILLIAC IV article in Issue 15 reminded some of our computers two (nonUIUC) of which CHM has in folks of their experiences. storage), and there is now an ILLIAC VI. I've also seen Former docent and old friend Bill Selmeier, writes from parts of the III and II at UIUC. Raleigh: • One really should not use the term "processors" they are more properly called processing elements (PEs) because When I was giving tours back in the old Visible they lacked control. Four control unit were planned, and Storage and got to the Illiac IV, I'd tell them this was CHM has the single surviving one. The PE memories my favorite exhibit. After explaining how the central were too small (barely 2 K words), so they were used program storage sent out the instructions to the 64 more like caches and the disk drives acted as primary midsized Burroughs processors to use on their own memory. data and how a bomb set off by radical students at It is more accurate to say that ILLIAC IV was designed the University of Wisconsin convinced the Dept. of • by Dan Slotnick and it was he that convinced both Defense to not let the world's most powerful computer Burroughs and the University of Illinois to team up to live minimally protected on any college campus, I'd build it. ask them if they liked science fiction. Did they remember a three part movie in the late 60's that • Keep in mind that a researcher was killed when the started off with some apes banging bones on the University of Wisconsin Madison was bombed. Two of ground. Of course most recognized 2001 A Space the three guys were caught and served time (now Odyssey, so then I'd ask how did Dave shut HAL released, interview on the web). The third has never down at the end. When they remembered that he surfaced. climbed inside the machine and began removing his • The project was picked up by Hans Mark, who at 38 was optical memory blocks, I'd lean over with a gloved the youngest director Ames ever had. He came from hand and pull out one of the boards in Illiac IV's main Lawrence Livermore Lab where he saw the store. A curator had told me that Illiac IV had been computational future when nuclear testing had to go shown to Kubrick when he was making the movie. He underground. My first Ames office actually had a copy of might have misunderstood and thought that all future the proposal which I wish I still had. computers would be built on that architecture. • "outperformed by the Cray" is not quite right. The ILLIAC, because it had fixed head disks, could do faster I/O (no Ed Thelen of the CHM 1401 Restoration project writes: seek time). Some jobs would be faster, others slower. I have two friends who were involved with Parallelism is constrained by the serial (Amdahl's Law). programming the ILLIAC IV at NASA. As • IVTRAN,TRANQUIL, and Glypnir were all dead programmers they might not have been so aware of languages, only good for academic papers. No one the troubles the hardware maintenance people were would use an ALGOLbased language in this country having. (When I was fixing computer equipment for except for toy problems. The only real operational G.E., I hoped the customers were (largely) unaware languages were CFD (CFD didn't stand for anything. I of the of the grief I was having trying to keep their asked my boss Ken who wrote it. Computation Fluid operations relatively smooth.) Dynamics was the principal application, but Ken denies I wanted to write a program for the ILLIAC IV. One of that was the reason for its name. If you believe that, care the (software) friends advised that I write it for only 40 to buy a bridge?) and Vectoral. Back then, only the processors, that way if some processors were down, University of Illinois (Dave Kuck) would do research on they could run it anyway. parallel language software. Dave has seen it all, I feel I have mentioned that suggested constraint to others, sorry somewhat for him. Dave has seen all the issues and was firmly told the machine was very reliable. and he has been ignored by many. Programming vector and parallel machines has been Eugene Miya • cited by DARPA as still a major problem. Don't let any • The fourth ILLIAC was not the last. The name continues one fool you. Only the simple stuff has been solved; to be used. An ILLIAC V was done, postnamed after the there is no simple “dropin” parallelism. real life Cedar Project (which consisted of many Alliant

Twentieth Anniversary of Linux Torvalds had wanted to call his invention Freax, from On Oct 4, 1991, Linus Benedict Torvalds' post to "freak", "free", and "x" (as an allusion to Unix). Early comp.os.minix begins the Linux OS. Finnish student, on,he stored the files under the name "Freax." Linus Torvalds announced his project to create an Torvalds had already considered the name "Linux," operating system kernel. Since then, the resulting but initially dismissed it as too egotistical.. One of Linux kernel has been marked by constant growth. It Torvald's coworkers did not think that "Freax" was a has grown from a small number of C files under a good name so he named the project "Linux" on the license prohibiting commercial distribution to its state server without consulting Torvalds. Later, however, in 2009 of over 370 megabytes of source under Torvalds consented to "Linux." Public License. Page 5 BABBAGE MACHINE REPAIRED Tim Robinson, with help from Randy Neff and Herb Kanner have replaced the broken bell crank and the Babbage engine can be demonstrated again. It is a temporary part but should hold up until the real thing arrives. Many thanks to Tim and all who helped.

Docent Quiz You didn't really expect answers, did you?

Coming Events Date Day Time Event Opening of the new exhibition: Oct 15 Sat. An Analog Life: Remembering Jim Williams 6:00 PM Reception 7:00 PM Worm: The First Digital World War. Author Mark Bowden and Oct 25 Tues. Microsoft's T.J. Campana in Conversation with John Markoff of The New York Times The Challenge and Promise of Artificial Intelligence, a Bay Area Nov 05 Sat. 02:00 PM Science Festival Wonder Dialog 6:00 PM Member Reception 7:00 PM The Technology of Animation DreamWorks Animation’s Jeffrey Nov 08 Tues. Katzenberg and Ed Leonard will kick off this series, in a conversation moderated by HP’s Phil McKinney. 6:00 PM Member Reception Nov 15 Tues. 7:00 PM A Computer Called Watson: IBM Research's David Ferrucci in Conversation with the Financial Times' Richard Waters

Please contribute to the Computer History Museum Volunteer Information Exchange. Share your stories, your interesting facts (and factoids) and your knowledge. Send them to Jim Strickland ([email protected])

Page 6